AI Made Video Cost Pennies. Where Did the Expensive Work Go?

Last week xAI opened its video model, Grok Imagine Video 1.5, to everyone (June 16). The numbers it published for itself are loud: it claims the top spot on the Image-to-Video Arena, a leaderboard where real people blind-rank which generated clip looks better, at roughly 86% below the price of OpenAI's Sora 2 Pro, with the soundtrack generated in the same pass as the picture so you never dub it separately.

Set the self-praise aside. The thing worth pausing on is the trend underneath it. A year ago, getting a decent short video with sound out of AI meant either paying a lot or stitching several tools together yourself: generate the picture, add the voice, fix the lip-sync. Now it is collapsing into a few-cents, one-step chore.

For someone who doesn't write code, the point here has nothing to do with the technology. The point is this: a step that used to cost money and time is turning almost free. A 60-second marketing video that once meant hiring a crew and paying by the day can now be batched out, a dozen candidates in an afternoon, by one person. I felt this building loomerce, my unattended e-commerce system: generating product clips went from 'needs a schedule and a budget' to 'click and move on.'

Here is the catch. Once 'making a video' costs almost nothing, it stops being your edge, because your competitor, your competitor's intern, and the bubble-tea shop down the street can all make one just as cheaply. There is a plain economic rule at work: the moment anyone can produce a thing, its price falls toward the cost of making it, and the money moves somewhere else.

So where did the expensive work go? It moved to two places the machine can't hold yet.

The first is knowing what to make. A model will spin up a hundred clips for you in an afternoon, but it doesn't know what your customer complained about yesterday, doesn't know which line makes a thumb stop scrolling, doesn't know that your brand can only talk like a human and can't fake it. The eye that judges which three of those hundred are worth posting, and why, goes by the word taste (here meaning the judgment of good from bad). This price cut didn't touch it. Output went up a hundredfold, and the eye that picks got more valuable, because now there are a hundred times as many things to pick from.

The second is getting seen. A finished video is only the starting line. Whether the right people surface it, whether they watch to the end, whether they act afterward, all of it goes by the word distribution (the ability to put content in front of the audience you want, which runs on channels, account weight, posting rhythm). AI pushed the production end to the floor, but it didn't win you a single extra second of attention. When everyone can make ten times more content, the total attention of platforms and viewers doesn't grow to match, so each piece gets a thinner slice than before.

What this price cut really does is lift the battlefield up a level. The old contest was 'can you make a video at all.' That bar is basically gone now, and the new contest is 'do you have something worth saying, and a channel that will carry it.' For an owner or a founder, that means reallocating budget and attention: stop agonizing over 'can we afford video,' that question is answered; spend the worry on building a channel that reliably reaches users, and on keeping a person who can judge what is worth making.

This is also my fixed move whenever the AI tools cycle through another update. For each new tool I ask one question first: which step did it just make cheap? Then immediately a second: so the people who used to earn a living on that step, what do they earn it on now? The headlines answer the first question for you. The second you have to think through yourself, and the answer is usually the valuable half.

Cheaper is good. It hands the right to make things to far more people, and that is real progress. Just remember that once anyone can do a thing, whether you can do it stops meaning anything. What is left is whether you have thought through what to make, and whether anyone will watch.